Abstracts

RATIONALE: The major cause of intractable temporal lobe epilepsy is mesial temporal sclerosis. The most frequent pathologic finding is selective neuronal hippocampal loss. However, pathology shows that other structures may be involved. We can name 1)temporal pole; 2)temporal lobe; 3)amygdala; 4)entorrhinal cortex/parahippocampal gyrus. 5)temporal horn of the lateral ventricle, and 6)hippocampus. We developed a protocol to measure the volumes of these structures based on MRI. The reliability was tested comparing measurements of patients to those of a paired control group. METHODS: The images were acquired at a 1.5 T Siemens Magneton unit. The volumetric images were reformatted in coronal plane, 2 mm thick, no slice gap, parallel to the main axis of the hippocampus. The structures were defined according to the anatomical landmarks published in the literature and manually delineated using the Scion Image 1.62 software, beta version (Scion Corp., Frederick, MD, USA). The volumes were calculated as absolute values and as ratios to the supratentorial volume (VST), as defined by our protocol. Percentage of right to left asymmetry was also calculated for each structure. Thirty (n = 30) temporal lobe patients and fifteen (n = 15) paired controls were analyzed. Statistical analysis was made for each structure (right and left) and for asymmetry index, comparing controls and patients. RESULTS: The results of the temporal horn of the lateral ventricles did not show significant difference between patients and controls. All other structures allowed distinction between the groups, with different levels of significance. The most significant difference was found in the hippocampal formation, followed by temporal pole, amygdala, entorrhinal cortex/parahippocampal gyrus and temporal lobe. CONCLUSIONS: With exception to the temporal horn, temporal lobe structures showed significant correlation to temporal lobe epilepsy. The different levels of significance may reflect different involvement in epileptogenesis. We are currently working on correlation with: a)postsurgical outcome; b)clinical features; c)EEG findings, and d)neuropsychological aspects.